Interpreting CFSE data
Fluorescent division tracking with CFSE is a powerful technique introduced by Lyons and Parish in 1994. Using this technique cell proliferation is followed by the development of peaks of cells found in different divisions. By harvesting cells at different times a great deal of information can be acquired about the dynamic properties of the responding cells. However, to do this one needs quantitative methods and, ideally, an accurate quantitative model of lymphocyte proliferation. We have developed a series of approaches, ranging from simple graphical methods to complex computational tools that can be used to analyse proliferation data to extract kinetic parameters. Our most recent models view lymphocytes as being comprised of independent machines controlling times to divide and times to die. Our experiments indicate that there is always a high degree of variation associated with these times and they are not typically inherited by daughter cells. Our probabilisitic model built on these principles can take time series data and find optimal parameters. The principles underlying this model apply to mouse T, B and human T and B lymphocyte proliferation. Thus, we have developed a useful generic model for exploring the effect of receptor delivered signals, drugs and genetic polymorphisms on immune outcomes.
The model also serves as a simulator of the immune response encompassing proliferation, cessation and subsequent death. We are now working to integrate the models for proliferation and survival with our understanding of differentiation to produce useful predictive tools. One application of these tools is described below.